Abstract: A method for image search, an electronic device, and a non-transitory computer-readable storage medium are provided. The method includes the following. Receive a query text. Process the query text to obtain a first multi-dimensional word vector. Perform, with a text gated network in a dual gate network, a first weighted operation on the first multi-dimensional word vector to obtain a second multi-dimensional word vector. Search for at least one target image according to the second multi-dimensional word vector and a second multi-dimensional visual vector for each image in an image file. The second multi-dimensional visual vector for each image is obtained by performing, with a visual gated network in the dual gate network, a second weighted operation on a first multi-dimensional visual vector for each image, and the second multi-dimensional word vector and the second multi-dimensional visual vector for each image are in a same space.

Abstract: Methods, systems, and apparatuses for image segmentation are provided. For example, a computing device may obtain an image, and may apply a process to the image to generate input image feature data and input image segmentation data. Further, the computing device may obtain reference image feature data and reference image classification data for a plurality of reference images. The computing device may generate reference image segmentation data based on the reference image feature data, the reference image classification data, and the input image feature data. The computing device may further blend the input image segmentation data and the reference image segmentation data to generate blended image segmentation data. The computing device may store the blended image segmentation data within a data repository. In some examples, the computing device provides the blended image segmentation data for display.

Abstract: Systems and methods for cardiac radioablation treatment planning are disclosed. In some examples, a computing device receives an image volume and a dose matrix data. The computing device generates a first mesh of organ substructures for the organ based on substructure contours for the organ. Further, the computing device generates a second mesh of an isodose volume based on the dose matrix data. The computing device displays the first mesh of the organ substructures and the second mesh of the isodose volume within a same scene. In some examples, the computing device samples a plurality of dosage values, determines representative dosage values for each of a plurality of points along surfaces of a dose matrix, and generates an image for display based on the representative dosage values. In some examples, a segmentation model is generated for display based on the representative dosage values.

Abstract: Technologies are presented herein in support of a system and method for performing fingerprint recognition. Embodiments of the present invention concern a system and method for capturing a user's biometric features and generating an identifier characterizing the user's biometric features using a mobile device such as a smartphone. The biometric identifier is generated using imagery captured of a plurality of fingers of a user for the purposes of authenticating/identifying the user according to the captured biometrics and determining the user's liveness. The present disclosure also describes additional techniques for preventing erroneous authentication caused by spoofing. In some examples, the anti-spoofing techniques may include capturing one or more images of a user's fingers and analyzing the captured images for indications of liveness.

Abstract: A method includes obtaining, by a processing device from an optical sensor of a mobile device, an image of an object. The method further includes processing, by the processing device, the image to identify the object in the image. Processing the image includes calculating at least one of a distance map or a reflection pattern. Processing the image further includes comparing the calculated distance map or the calculated reflection pattern with a known distance map or a known reflection pattern to determine whether the image contains a spoof or a real object. The method further includes obtaining, by the processing device from the image, data including a biometric characteristic of the user. The method further includes sending, by the processing device to a third party computing device, data including at least the biometric characteristic.

Abstract: Access control system and method to distinguish between tailgate and piggyback. The method includes receiving, from a camera positioned at an access point, one or more images of a first person at the access point and determining based on credential information whether the first person is authorized to enter through the access point. In response to determining that the first person is authorized to enter through the access point, the method includes determining, based on a consent-to-follow gesture of the first person detected in the one or more images, a number of guests of the first person requested to enter through the access point. The method also includes determining, based on the one or more images, a number of persons entering through the access point and generating an alert indicating unauthorized entry when the number of persons is greater than the number of guests.

Abstract: Systems and methods are described for generating pixel image data, using a lensless camera, based on light that travels through a mask that with pattern masking the lensless camera. The system applies a transformation function to the pixel image data to generate frequency domain image data. The system inputs the frequency domain image data into a machine learning model, wherein the machine learning model does not have access to data that represents the pattern of the mask. The model is trained using a set of images with the feature that are captured by the flat, lensless camera through the mask. The system processes the frequency domain image data using the machine learning model to determine whether the pixel image data depicts the image feature. The system further performs an action based on determining that the pixel image data depicts the image feature.

Abstract: Methods, devices, and a non-transitory computer-readable storage medium are provided for determining a video cover image. The method includes: obtaining a candidate image set containing a plurality of image frames to be processed by determining the plurality of image frames to be processed from a video to be processed, each of the plurality of image frames to be processed containing at least one target object; obtaining a target score of each of the plurality of image frames to be processed by inputting the plurality of image frames to be processed in the candidate image set into a scoring network; and sorting the target scores of the plurality of image frames to be processed according to a set order to obtain a sorting result, and determining a video cover image of the video to be processed from the plurality of image frames to be processed according to the sorting result.

Abstract: The subject invention pertains to methods and systems for classifying leukocytes using artificial intelligence called AIRFIHA (artificial-intelligence enabled reagent-free imaging hematology analyzer) that can accurately classify subpopulations of leukocytes in a label-free manner. AIRFIHA can not only subtype lymphocytes into B and T cell but is capable of sorting different types of T cells subtypes. AIRFIHA is realized through training a two-step neural network using label-free images of separated leukocytes acquired from a custom-built quantitative phase microscope. Owing to its easy operation, low cost, and strong discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource-limited settings.

Abstract: Identifying an indistinct entity within an image can include generating by an image filter multiple gradients, each of which corresponds to one of a plurality of pixels of an image captured by an imager. The image can be searched for a likely repeating pattern. Responsive to detecting, based on the multiple gradients, a likely repeating pattern within the image, data structures can be generated, the data structures comprising a set of probabilistically weighted feature vectors corresponding to the likely repeating pattern. A machine learning model can classify each of the set of probabilistically weighted feature vectors. An identity of the likely repeating pattern can be output, the identity based on the machine learning model classifications of the probabilistically weighted feature vectors.

Abstract: A computer-implemented method of segmenting a reconstructed volume of a region of patient anatomy includes: determining an anatomical region associated with the reconstructed volume; detecting one or more metal objects disposed in an initial 3D metal object mask associated with the reconstructed volume; for each of the one or more metal objects disposed in the initial 3D metal object mask, determining a volume associated with the metal object; determining a value for at least one segmentation parameter based on the anatomical region and on the volume associated with the one or more metal objects; and generating a final 3D metal object mask associated with the reconstructed digital volume using the value for the segmentation parameter.

Abstract: Technologies are presented herein in support of a system and method for performing fingerprint recognition. Embodiments of the present invention concern a system and method for capturing a user's biometric features and generating an identifier characterizing the user's biometric features using a mobile device such as a smartphone. The biometric identifier is generated using imagery captured of a plurality of fingers of a user for the purposes of authenticating/identifying the user according to the captured biometrics and determining the user's liveness. The present disclosure also describes additional techniques for preventing erroneous authentication caused by spoofing. In some examples, the anti-spoofing techniques may include capturing one or more images of a user's fingers and analyzing the captured images for indications of liveness.

Abstract: Head mountable systems, methods, and non-transitory computer readable media including instructions for identifying individuals using facial skin micromovements are disclosed. An example head mountable system may include a wearable housing, a coherent light source, a detector, and at least one processor. The at least one processor may analyze reflection signals from the detector to determine specific facial skin micromovements of an individual wearing the head mountable system. Thereafter, the at least one processor may access memory correlating a plurality of facial skin micromovements with the individual, and search for a match between the determined specific facial skin micromovements and at least one of the plurality of facial skin micromovements in the memory. If a match is identified, the processor may initiate a first action; and if a match is not identified, the at least one processor may initiate a second action different from the first action.

Abstract: An information processing apparatus according to an embodiment of the present technology includes a classification unit and a generation unit. The classification unit classifies an object detected in a space on a basis of a predetermined criterion. The generation unit sets a priority for the object on a basis of a classification result by the classification unit, and generates position-related information regarding a position in the space on a basis of the set priority. Use of the position-related information makes it possible to improve the accuracy of autonomous movement control. This makes it possible to improve the accuracy of autonomous movement control.

Abstract: Provided are an image processing device, an image processing method, and a program for recognizing an object in a three-dimensional map, which do not require collecting learning data of the three-dimensional map and can perform high-speed processing with a small load. The image processing device includes an image acquiring section that sequentially acquires a two-dimensional input image for each frame; an object type recognition executing section that attaches, to each of pixels of the input image acquired for each frame, a label indicating a type of an object represented by the pixels; and a labeling section that executes three-dimensional position recognition of a subject represented in the input image to create a three-dimensional map, based on the input image sequentially input, and attaches, to each voxel included in the three-dimensional map, the label of the pixel corresponding to the voxel.

Abstract: The embodiments of the present disclosure provide a method and an apparatus for acquiring a RAW image, and an electronic device, the method including: obtaining an sRGB image; inputting the sRGB image into a preset image restoration model to obtain an initial RAW image corresponding to the sRGB image, wherein the preset image restoration model is a model obtained by training a preset neural network based on a first preset training set; adding a first preset noise to the initial RAW image to obtain a first noise RAW image; adding a second preset noise to the initial RAW image to obtain a second noise RAW image; inputting the first noise RAW image and the second noise RAW image into a preset noise addition model to obtain a target RAW image, wherein the preset noise addition model is a model obtained by training a preset neural network based on a second preset training set.

Abstract: A reinforcement learning based approach to the problem of query object localization, where an agent is trained to localize objects of interest specified by a small exemplary set. We learn a transferable reward signal formulated using the exemplary set by ordinal metric learning. It enables test-time policy adaptation to new environments where the reward signals are not readily available, and thus outperforms fine-tuning approaches that are limited to annotated images. In addition, the transferable reward allows repurposing of the trained agent for new tasks, such as annotation refinement, or selective localization from multiple common objects across a set of images. Experiments on corrupted MNIST dataset and CU-Birds dataset demonstrate the effectiveness of our approach.

Abstract: A pixel is appropriately interpolated regarding a histogram image in which a distance histogram is allocated to each pixel position. A signal processing apparatus according to the present technology includes a histogram generation unit that inputs a histogram image in which a distance histogram indicating results of a plurality of times of distance measurement as frequency information for each distance is allocated to each pixel position and generates a distance histogram of an interpolation target position in the histogram image on the basis of distance histograms of a plurality of pixel positions near the interpolation target position.

Abstract: Aspects of the technology described herein improve an object recognition system by specifying a type of picture that would improve the accuracy of the object recognition system if used to retrain the object recognition system. The technology described herein can take the form of an improvement model that improves an object recognition model by suggesting the types of training images that would improve the object recognition model's performance. For example, the improvement model could suggest that a picture of a person smiling be used to retrain the object recognition system. Once trained, the improvement model can be used to estimate a performance score for an image recognition model given the set characteristics of a set of training of images. The improvement model can then select a feature of an image, which if added to the training set, would cause a meaningful increase in the recognition system's performance.

Abstract: A 3D segmentation editing system accurately updates the segmentations of non-edited images of a 3D scan to reflect segmentation edits applied to other images of the scan using localized interpolation. In one or more embodiments, rather than replacing the entireties of the initial segmentations of non-edited images with newly generated, globally interpolated segmentations, the segmentation editing system applies a distance-based criterion to the interpolation of segmentation edits, such that only portions of the segmentations of the non-edited images that correspond to areas that were manually annotated in the edited images will be modified by the interpolation process, and the initial segmentations will be maintained outside of those edited areas. In this way, the system merges the interpolated segmentation with the initial segmentation for each non-edited image in a manner that mitigates unreliable modifications to the initial segmentations in areas far from the edited areas.